Certain manifestations of Lyme disease (LD) do not resolve with antibiotic therapy. We have studied this problem previously in patients with Lyme arthritis who have persistent synovitis for months or years after spirochetal killing with antibiotics, called antibiotic-refractory Lyme arthritis. We have hypothesized that this disease course results from infection-induced autoimmunity with immune dysregulation in joints. However, until recently, identification of pathogenic autoantigens has remained elusive. Using discovery-based proteomics combined with clinical translational research, we recently identified endothelial cell growth factor (ECGF) as the first known autoantigen that induces T and B cell responses in patients with antibiotic- refractory Lyme arthritis. Our first specific aim in this application is to determine the clinical correlates of autoimmunity across the spectrum of LD, including erythema migrans, neuroborreliosis, carditis, arthritis, and post-infectious syndromes, to assess whether this response is specific for LD, and to learn whether it occurs only with U.S. B. burgdorferi (Bb) strains or also with European genospecies of the spirochete. Second, we want to understand spirochetal and host cell interactions, including the types of Bb strains and host cell types, which are necessary for the induction of ECGF and ECGF immune responses. Third, we plan to characterize the phenotype, function and gene expression of immune cell subsets that lead to immune dysregulation in joints in patients with antibiotic-refractory arthriti, and to learn whether such abnormalities occur in patients with other LD pictures. These aims will allow us to expand our observation that autoimmunity to ECGF can occur as a part of the spectrum of Bb-induced immune responses, to ascertain whether specific clinical pictures result from such immune responses, to delineate mechanisms that account for these responses, and to characterize features of immune dysregulation that lead to persistent disease activity. If these studies validate ECGF as a pathogenic autoantigen, we anticipate that a test for this antibody response will become a part of the clinical evaluation of LD. Finally, this work has implications beyond LD, since it may create a new paradigm for how to identify and understand infection-induced autoimmunity with immune dysregulation in other diseases.